Computer-Assisted Segmentation of White Matter Lesions in 3D MR Images Using Support Vector Machine

Zhiqiang Lao, Dinggang Shen, Dengfeng Liu, Abbas F. Jawad, Elias R. Melhem, Lenore J. Launer, R. Nick Bryan, Christos Davatzikos

Research output: Contribution to journalArticle

Abstract

Rationale and Objectives: Brain lesions, especially white matter lesions (WMLs), are associated with cardiac and vascular disease, but also with normal aging. Quantitative analysis of WML in large clinical trials is becoming more and more important. Materials and Methods: In this article, we present a computer-assisted WML segmentation method, based on local features extracted from multiparametric magnetic resonance imaging (MRI) sequences (ie, T1-weighted, T2-weighted, proton density-weighted, and fluid attenuation inversion recovery MRI scans). A support vector machine classifier is first trained on expert-defined WMLs, and is then used to classify new scans. Results: Postprocessing analysis further reduces false positives by using anatomic knowledge and measures of distance from the training set. Conclusions: Cross-validation on a population of 35 patients from three different imaging sites with WMLs of varying sizes, shapes, and locations tests the robustness and accuracy of the proposed segmentation method, compared with the manual segmentation results from two experienced neuroradiologists.

Original languageEnglish (US)
Pages (from-to)300-313
Number of pages14
JournalAcademic Radiology
Volume15
Issue number3
DOIs
StatePublished - Mar 2008
Externally publishedYes

Fingerprint

Magnetic Resonance Imaging
Vascular Diseases
Protons
Heart Diseases
Support Vector Machine
White Matter
Clinical Trials
Brain
Population

Keywords

  • machine learning
  • support vector machine
  • White matter lesion segmentation

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Lao, Z., Shen, D., Liu, D., Jawad, A. F., Melhem, E. R., Launer, L. J., ... Davatzikos, C. (2008). Computer-Assisted Segmentation of White Matter Lesions in 3D MR Images Using Support Vector Machine. Academic Radiology, 15(3), 300-313. https://doi.org/10.1016/j.acra.2007.10.012

Computer-Assisted Segmentation of White Matter Lesions in 3D MR Images Using Support Vector Machine. / Lao, Zhiqiang; Shen, Dinggang; Liu, Dengfeng; Jawad, Abbas F.; Melhem, Elias R.; Launer, Lenore J.; Bryan, R. Nick; Davatzikos, Christos.

In: Academic Radiology, Vol. 15, No. 3, 03.2008, p. 300-313.

Research output: Contribution to journalArticle

Lao, Z, Shen, D, Liu, D, Jawad, AF, Melhem, ER, Launer, LJ, Bryan, RN & Davatzikos, C 2008, 'Computer-Assisted Segmentation of White Matter Lesions in 3D MR Images Using Support Vector Machine', Academic Radiology, vol. 15, no. 3, pp. 300-313. https://doi.org/10.1016/j.acra.2007.10.012
Lao, Zhiqiang ; Shen, Dinggang ; Liu, Dengfeng ; Jawad, Abbas F. ; Melhem, Elias R. ; Launer, Lenore J. ; Bryan, R. Nick ; Davatzikos, Christos. / Computer-Assisted Segmentation of White Matter Lesions in 3D MR Images Using Support Vector Machine. In: Academic Radiology. 2008 ; Vol. 15, No. 3. pp. 300-313.
@article{9454f6a289de474eaf89c9d92e9dcc7f,
title = "Computer-Assisted Segmentation of White Matter Lesions in 3D MR Images Using Support Vector Machine",
abstract = "Rationale and Objectives: Brain lesions, especially white matter lesions (WMLs), are associated with cardiac and vascular disease, but also with normal aging. Quantitative analysis of WML in large clinical trials is becoming more and more important. Materials and Methods: In this article, we present a computer-assisted WML segmentation method, based on local features extracted from multiparametric magnetic resonance imaging (MRI) sequences (ie, T1-weighted, T2-weighted, proton density-weighted, and fluid attenuation inversion recovery MRI scans). A support vector machine classifier is first trained on expert-defined WMLs, and is then used to classify new scans. Results: Postprocessing analysis further reduces false positives by using anatomic knowledge and measures of distance from the training set. Conclusions: Cross-validation on a population of 35 patients from three different imaging sites with WMLs of varying sizes, shapes, and locations tests the robustness and accuracy of the proposed segmentation method, compared with the manual segmentation results from two experienced neuroradiologists.",
keywords = "machine learning, support vector machine, White matter lesion segmentation",
author = "Zhiqiang Lao and Dinggang Shen and Dengfeng Liu and Jawad, {Abbas F.} and Melhem, {Elias R.} and Launer, {Lenore J.} and Bryan, {R. Nick} and Christos Davatzikos",
year = "2008",
month = "3",
doi = "10.1016/j.acra.2007.10.012",
language = "English (US)",
volume = "15",
pages = "300--313",
journal = "Academic Radiology",
issn = "1076-6332",
publisher = "Elsevier USA",
number = "3",

}

TY - JOUR

T1 - Computer-Assisted Segmentation of White Matter Lesions in 3D MR Images Using Support Vector Machine

AU - Lao, Zhiqiang

AU - Shen, Dinggang

AU - Liu, Dengfeng

AU - Jawad, Abbas F.

AU - Melhem, Elias R.

AU - Launer, Lenore J.

AU - Bryan, R. Nick

AU - Davatzikos, Christos

PY - 2008/3

Y1 - 2008/3

N2 - Rationale and Objectives: Brain lesions, especially white matter lesions (WMLs), are associated with cardiac and vascular disease, but also with normal aging. Quantitative analysis of WML in large clinical trials is becoming more and more important. Materials and Methods: In this article, we present a computer-assisted WML segmentation method, based on local features extracted from multiparametric magnetic resonance imaging (MRI) sequences (ie, T1-weighted, T2-weighted, proton density-weighted, and fluid attenuation inversion recovery MRI scans). A support vector machine classifier is first trained on expert-defined WMLs, and is then used to classify new scans. Results: Postprocessing analysis further reduces false positives by using anatomic knowledge and measures of distance from the training set. Conclusions: Cross-validation on a population of 35 patients from three different imaging sites with WMLs of varying sizes, shapes, and locations tests the robustness and accuracy of the proposed segmentation method, compared with the manual segmentation results from two experienced neuroradiologists.

AB - Rationale and Objectives: Brain lesions, especially white matter lesions (WMLs), are associated with cardiac and vascular disease, but also with normal aging. Quantitative analysis of WML in large clinical trials is becoming more and more important. Materials and Methods: In this article, we present a computer-assisted WML segmentation method, based on local features extracted from multiparametric magnetic resonance imaging (MRI) sequences (ie, T1-weighted, T2-weighted, proton density-weighted, and fluid attenuation inversion recovery MRI scans). A support vector machine classifier is first trained on expert-defined WMLs, and is then used to classify new scans. Results: Postprocessing analysis further reduces false positives by using anatomic knowledge and measures of distance from the training set. Conclusions: Cross-validation on a population of 35 patients from three different imaging sites with WMLs of varying sizes, shapes, and locations tests the robustness and accuracy of the proposed segmentation method, compared with the manual segmentation results from two experienced neuroradiologists.

KW - machine learning

KW - support vector machine

KW - White matter lesion segmentation

UR - http://www.scopus.com/inward/record.url?scp=39049098162&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=39049098162&partnerID=8YFLogxK

U2 - 10.1016/j.acra.2007.10.012

DO - 10.1016/j.acra.2007.10.012

M3 - Article

C2 - 18280928

AN - SCOPUS:39049098162

VL - 15

SP - 300

EP - 313

JO - Academic Radiology

JF - Academic Radiology

SN - 1076-6332

IS - 3

ER -